ToPy Topology Optimization Framework

You can define a problem directly in Python using a configuration dictionary. This is useful for rapid experimentation without modifying external text files. The dictionary keys and requirements are identical to those used in a .tpd file.

Pass the dictionary to the Topology constructor via the config argument.

config = {
     'DOF_PN': 2,
     'ELEM_K': 'Q4',
     'ETA': '0.5',
     'FILT_RAD': 1.5,
     'FXTR_NODE_X': range(1, 22),
     'FXTR_NODE_Y': 1281,
     'LOAD_NODE_Y': 1,
     'LOAD_VALU_Y': -1,
     'NUM_ELEM_X': 60,
     'NUM_ELEM_Y': 20,
     'NUM_ELEM_Z': 0,
     'NUM_ITER': 94,
     'PROB_NAME': 'beam_2d_reci',
     'PROB_TYPE': 'comp',
     'P_FAC': 3.0,
     'VOL_FRAC': 0.5
}

from topy import Topology
topology = Topology(config=config)

The first time you run ToPy, it may automatically generate element stiffness matrices using SymPy. This process can take several minutes and is usually only required once.

If you want to manually trigger the creation of these matrices without solving a full problem, run the optimise.py script located in the scripts folder.

# Run from the scripts directory to generate matrices
python scripts/optimise.py

To solve a problem defined in a TPD (ToPy Problem Definition) file, use the optimise.py script. You must run this command from within the specific example directory containing the .tpd file.

Note for Windows users: Unlike Linux or macOS, you cannot use symbolic links to reference a central optimise.py script. You must manually copy the optimise.py file into each example directory where you wish to run it.

python optimise.py <filename>.tpd

If you are unsure whether your Python installation is 32-bit or 64-bit (which is critical for selecting the correct Windows binaries), run the following command in a Python shell:

import platform
platform.architecture()

ToPy can be installed from source after downloading its dependencies.

Note on Python Versions:

• The stable release (v0.4.1) is designed for Python 2 and works with Pysparse (recommended if using the Anaconda distribution).
• The latest master branch (unstable) supports Python 3.

To install the stable release:

1. Download the latest stable release from the GitHub releases page.
2. Navigate to the topy/topy directory.
3. Run the installation command.

$ cd topy/topy
$ python setup.py install

The topy.visualization module allows you to export results for viewing:

• 2D problems: Save as .png images.
• 3D problems: Save as .vtk files (viewable with Mayavi or ParaView).

To create an animation from a sequence of PNG images, you can use the ImageMagick convert tool.

convert -delay 35 *.png anim.gif

You can solve a defined problem using either the command line or a Python script.

Command Line

Run the optimise.py script and pass the path to your .tpd file:

python topy/scripts/optimise.py <filename>.tpd

Python API

1. Initialize a Topology object with your config.
2. Call .set_top_params() to prepare the instance.
3. Call topy.optimise(t) to run the solver.

import topy

config = {...}
t = topy.Topology(config)
t.set_top_params()
topy.optimise(t)

If you have a sequence of PNG images representing an animation, you can use ImageMagick's convert tool to compile them into a single GIF. This is useful for visualizing ToPy results.

convert -delay 35 *.png anim.gif

The optimise.py script is a CLI utility used to read a ToPy Problem Definition (TPD) file and execute the optimization process to solve the defined problem.

python scripts/optimise.py <path_to_tpd_file>